Would you rather talk with Alexa or Samuel L. Jackson? At the Amazon Devices Event in Seattle on Wednesday, Amazon announced celebrity voices for Alexa in addition to a full array of smart devices.
According to Dave Limp, Amazon senior vice president of Devices, “When we introduced Alexa, we used state-of-the-art text to speech technology. Now, we’re working to make Alexa’s personalities — and customer interactions more natural.”
Amazon uses neural text-to-speech (TTS) to invent new ways for Alexa (and celebrities) to interact with Echo device owners.
“While customers love Alexa’s voice, they sometimes want to hear a completely different voice. Customers loved getting morning greetings from Chelsea Handler and a duet with Ed Sheeran,” Limp said.
“In the past, we had to script and record all of these interactions,” Limp continued, “but with neural text to speech, we’re able to model a celebrity voice directly.”
Amazon chose iconic actor and producer Samuel L. Jackson as the first celebrity voice for Alexa. “Jackson can tell you jokes, let you know if it’s raining, set timers and alarms, play music and more — all with a bit of his own personality.”
You will still need to address Alexa to hear a response in Jackson’s voice. For example, you could say:
-
- “Alexa, ask Samuel L. Jackson where he is from?”
- “Alexa, ask Samuel L. Jackson to wake me up at 7 A.M.”
- “Alexa, ask Samuel L. Jackson to sing happy birthday.”
There will be two versions of Jackson’s voice — one will use explicit language and the other will be non-explicit. Jackson’s voice will be available for interaction later this year. Customers who add Jackson’s personality will be able to do so for just $1 during the introduction this year. After the introductory period the price will rise to $5. Additional celebrity voices for Alexa will be available in 2020.
Alexa (or Jackson) will also be able to adjust to your conversational patterns. Amazon built a deep-learning model that detects when customers are frustrated with Alexa. When you’re upset with Alexa — and not the world in general — the artificial intelligence model will be able to adapt on the fly. Frustration adaptation is separate from the celebrity voice feature.
